پیشرفت های اخیر در درمان لکنت زبان: چشم انداز نظری

Prolonged speech and its variants are widely used in the behavioral treatment of stuttering. Unlike these approaches, which depend on clinician-prescribed speech pattern changes, two behavioral treatment regimens, one for children and another for adults, recently developed at the Australian Stuttering Research Center, promote self-monitoring of speech as a means of controlling stuttering. In these programs, the clients themselves modify their speech in subtle and variable ways to gain control over stuttering and, in that, they appear to be similar to a well-known experimental technique for suppressing stutters known as response contingent stimulation. The present paper provides an integrated explanation for the effectiveness of both clinician-directed as well as client-initiated speech pattern modifications and, in the process, develops a new model of stuttering. It also shows why client-generated speech patterns changes potentially produce faster and more lasting improvement than those changes prescribed by a clinician.

For over half a century, primarily influenced by Johnson's (1942) diagnosogenic theory, speech–language pathologists had shown great trepidations about offering direct treatment for stuttering in young children. Recently, however, work carried out at the Australian Stuttering Research Center, which is based on a path-breaking study reported by Martin, Kuhl, and Harlodson (1972), has shown that stuttering in children can be treated safely (Woods, Shearsby, Onslow, & Burnham, 2002), efficaciously (Harris, Onslow, Packman, Harrison, & Menzies, 2002; Onslow, Menzies, & Packman, 2001), and economically (Jones, Onslow, Harrison, & Packman, 2000) with a home-based operant treatment regimen called the Lidcombe program. In the Lidcombe program, parents are trained to: (1) administer verbal praise and occasionally tangible rewards contingent on fluent utterances; (2) request in a nonthreatening manner that stuttered utterances be replaced with fluent utterances; and (3) rate stuttering severity on a daily basis on a 10-point scale (Onslow et al., 2001a and Onslow et al., 2001b). The most striking feature of the Lidcombe program is that neither the clinicians who work with the child while training the parents in the clinic nor parents who, for the most part, administer the treatment outside of the clinic attempt to modify the child's speech either through instructions (e.g., “speak slowly” or “draw out words,” etc.) or by way of modeling. Onslow, Stocker, Packman, and McLeod (2002), after failing to find consistent acoustic timing differences between pre- and post-treatment speech of a group of children who successfully completed the Lidcombe program, concluded that no satisfactory explanation exists for the success of the program. In another study, O’Brian, Onslow, Cream, and Packman (2003) described a treatment regimen for adults called the Camperdown program that initially and briefly required participants to adopt a slow (70 syllables/min), prolonged speech pattern modeled in a videotape. In later stages of treatment, participants were only required to produce speech that was rated 1–2 on a 9-point stuttering severity scale and 1–3 on a 9-point speech naturalness scale without having to meet any specific targets for speech modification such as speech rate, gentle voice onset, continuous vocalization, etc. Sixteen participants, out of the original group of 30, who completed the program met and maintained the stuttering severity and speech naturalness criteria and achieved a satisfactory rating on a lay listener based social validation measure 12 months post-treatment. Prolonged speech based stuttering treatment typically involves “shaping” speech systematically by requiring participants to meet specific criteria for rate and an assortment of related speech modifications such as gentle voice onset, continuous vocalization, and soft articulatory contact in small, incremental steps (Ingham, 1984 and Onslow, 1996). The Camperdown program demonstrated that people who stutter (PWS) could develop natural-sounding, nearly stutter-free speech without specific clinician instructions with regard to speech modifications although they do require consistent and reliable feedback concerning stuttering severity and speech naturalness. The success of the Lidcombe and the Camperdown programs appear to suggest that many children and adults who stutter are able to produce nearly stutter-free and natural-sounding speech by: (1) developing a cognitive set to speak without stutters and (2) monitoring their speech, initially with the help of clinicians or family members, to verify that this goal is achieved. This inference has strong empirical foundation. Numerous studies have demonstrated that response contingent stimulation (RCS) – presenting almost any kind of “stimulus” (more descriptively, any kind of signal [ Wingate, 1980]) immediately and consistently contingent on stutters – significantly reduces or eliminates stutters in most but not all PWS (see Bloodstein, 1995; Costello & Ingham, 1984; Ingham, 1984, and Prins & Hubbard, 1988 for reviews). Although initially this finding was interpreted as evidence that stutters were an operant response class, the failure to increase stutter frequency through positive or negative reinforcement ( Bloodstein, 1995; Daly & Kimbarow, 1978; Young, 1985) has generally led to the abandonment of that view ( Ingham, 1984). The perplexing finding that almost any kind of signal – positive, negative, and seemingly neutral – when paired with stutters would produce measurable decreases in stutter frequency in many PWS has been attributed by some to a “highlighting” effect (Siegel, 1970 and Wingate, 1980). Siegel maintained that “… virtually any event that highlights or brings (stutters) to the speaker's attention …” (p. 689) will reduce the frequency of stutters. In fact, when some PWS highlight their own stutters by some means such as by pressing a handswitch every time they stutter, stutter frequency is reduced (Hanson, 1978). James, Ricciardelli, Rogers, and Hunter (1989) suggested that when stutters are highlighted systematically as in RCS studies, the PWS might become more fluent by tapping fluent speech capabilities that remained unused in “contingency-free” speaking conditions. However, James et al. did not identify the origin and nature of “fluent speech capabilities” that are underutilized by PWS. If PWS, in RCS experiments as well as in stuttering treatment programs derived from operant learning principles, are accessing normally unexploited fluent speech capabilities, it is important to identify the source(s) of these hidden capabilities. Recently, a number of researchers have proposed that the “stage” of speech production that is of direct relevance to an explanation of stuttering is the speech motor plan assembly and its execution (Peters, Hulstijn, & Van Lieshout, 2000; Postma, Kolk, & Povel, 1990; Wijnen & Boers, 1994). The present paper offers a speech motor plan assembly explanation for the suppression of stutters under novel speech patterns (Andrews, Howie, Dozsa, & Guitar, 1982) including those associated with the behavioral treatment of stuttering that systematically promote self-monitoring of speech by PWS.

Prolonged speech and its variants constitute the most common and arguably the most successful therapy offered for stuttering (Onslow, 1996). Nearly all extant explanations of ameliorative effect of novel speech patterns, of which prolonged speech is just one of many possibilities, rests on the assumption that the novel speech patterns somehow simplify the speech production process (Adams, 1990; Bloodstein, 1995; Perkins, Kent, & Curlee, 1991). However, in spite of intense laboratory research, support for these explanations is scant (Ingham, 1984; Onslow, 1996). In this paper, I have argued that a resource intensive, foreground process called speech construction is the basis for fluent speech in PWS who successfully complete operant learning based stuttering treatment regimens. In addition, the present paper provides a rationale as to why cognitively driven speech construction – speaking while consciously monitoring one's speech to achieve a certain communication goal which, in this case, is speaking without stutters – produces faster and superior results than motorically driven speech construction – speaking while consciously altering certain specific aspects of articulation and prosody. Finally, empirical data appear to suggest that people who stutter are slow to retrieve stored motor plans, which leaves less and, in instances that produce stutters, insufficient time to retrieve the entire motor plan. Very young children who are at risk for chronic stuttering may be able to permanently overcome this hypothesized defect in speech motor plan assembly if placed in a program that systematically encourages them to monitor their speech for fluent utterances and stutters.